{"id":"W4408177377","doi":"10.1016/j.sbi.2025.103027","title":"Language models for protein design","year":2025,"lang":"en","type":"review","venue":"Current Opinion in Structural Biology","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Protein design; Computational biology; Computer science; Linguistics; Biology; Protein structure; Biochemistry; Philosophy","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001532206,0.001458076,0.001294459,0.00125137,0.0004159757,0.001808862,0.001859497,0.002361616,0.006343672],"category_scores_gemma":[0.002417389,0.0004326321,0.0009762658,0.001508041,0.002030044,0.003081909,0.001130151,0.003222158,0.004983629],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001684838,"about_ca_system_score_gemma":0.001866041,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0018541,"about_ca_topic_score_gemma":0.001333003,"domain_scores_codex":[0.9994716,0.0001904301,0.00003746392,0.00009923548,0.0001694011,0.00003192127],"domain_scores_gemma":[0.9991232,0.0006253408,0.00005120947,0.00006220069,0.000112523,0.00002558239],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00003793326,0.00006295571,0.0001690135,0.005323396,0.0001283043,0.0001519055,0.0001373273,0.01485852,0.001219176,0.4602265,0.04923505,0.46845],"study_design_scores_gemma":[0.00001728034,0.00003999656,0.0001377552,0.00153273,0.00003823572,0.0003523361,0.00004181191,0.008834697,0.0006818794,0.3524434,0.6358408,0.00003911806],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.0005494437,0.9039651,0.06891289,0.005908702,0.001352162,0.00004266942,0.0002234741,0.0002974376,0.01874806],"genre_scores_gemma":[0.01889429,0.9383048,0.02925854,0.00243256,0.001730738,0.0001570278,0.0004988659,0.00009727344,0.008625942],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.006343672,"threshold_uncertainty_score":0.02122164,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06558631815449587,"score_gpt":0.4139570759980702,"score_spread":0.3483707578435743,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}